1 CHE 322 : Enzyme Mechanisms and Natural Product Biosynthesis

Prof. J. A. Robinson

Contents

1. Enzyme Chemistry

1.1. How efficient are enzymes as catalysts?

1.2. Some mechanisms of chemical catalysis

1.3. Enzyme kinetics

1.4. General acid-base catalysis and nucleophilic/electrophilic catalysis in enzymic reactions

1.5. Redox-reactions and the coenzymes that mediate them.

2. Biosynthesis of Natural Products

2.1 Terpenes

2.2 Polyketides

2.3 Shikimic acid and phenyl-propanoide natural products

2.4 Alkaloids

Recommended Literature :

1) From Enzyme Models to Model Enzymes, A. J. Kirby and F. Hollfelder, RSC Publishing,

Camdridge, UK, 2009.

2) An Introduction to Enzyme and Coenzyme Chemistry, T. Bugg, Blackwell Science, 1997.

3) Medicinal Natural Products - A Biosynthetic Approach, P. M. Dewick, J. Wiley, 2002

4) The Organic Chemistry of Enzyme Catalyzed Reactions, R. B. Silverman, Academic Press, 2000.

5) Structure and Mechanism in Protein Science, A. R. Fersht, W. H. Freeman & Co., 1998.

6) Chemical Aspects of Biosynthesis, J. Mann, Oxford Science Publ., 1994.

7) Topics in Current Chemistry, Vol. 195 Biosynthesis: Polyketides and Vitamins; and Vol. 209

Biosynthesis: Aromatic Polyketides, Isoprenoids, Alkaloids. Springer Verlag, 1998 and 2000.

8) The Organic Chemistry of Biological Pathways, J. E. McMurry & T. P. Begley, Roberts & Co.

Publ., USA, 2005

The main goals of these lectures are :

- To provide an introduction into the mechanisms of enzymic reactions, and to introduce the most

important coenzymes.

- To provide a mechanistic understanding of how the different classes of complex natural products are

constructed in Nature.

- To illustrate the methods and techniques that are used to study biosynthetic pathways.

2 1. Enzyme Chemistry 1.1. How efficient are enzymes as catalysts? Essentially all chemical reactions that take place within cells are catalyzed by enzymes. It is, therefore, interesting to investigate how efficient enzymes are as catalysts, and what mechanisms they use to achieve rate enhancements. But how large are the rate enhancements achieved by enzymes? How much faster do enzymic reactions proceed, in comparison to similar non-enzymic processes. Is there an upper limit to the rate enhancements that can be achieved by enzymes, and if so what is this limit? We will attempt to answer these questions in the first part of this lecture course. But quite often this is not so easy to do, because in the absence of an enzyme a reaction might not proceed at a measureable rate at all. E.g. Hydrolysis of a peptide bond (-CO-NH-):

k (150oC) = 6 x 10-7 s-1 ; from the temperature dependence of the rate constant we can estimate that knon (RT) ≈ 4 x 10-11 s-1 ; A first order reaction in water at pH 7 (J. Am. Chem. Soc., 1996, 118, 5498 und 6105), with a half-life of about 500 years at room temperature (RT). In the case of peptide hydrolysis catalyzed by a typical protease (Carboxypeptidase B): Peptide = ...Gly.Gly..... kcat ≈ 240 s-1 (kcat = number of substrate molecules converted to product, per enzyme molecule (or active site), per second). This gives kcat/knon ~ 1013 (see JACS 1996, 118, 6105; compare JACS 2005, 127, 10828; and JACS 2011, 133, 13821). Example 2. (J. Am. Chem. Soc. 1999, 121, 11831):

Example 3. ((Science, 1995, 267, 90) :

knon (pH independent; 1st order reaction) = 2.8 x 10-16 sec-1 kcat (yeast OMP-decarboxylase) = 39 s-1 In this case the reaction rate is increased by the enzyme by a factor of ≈ 1017 !

AcNH

HN CO-NHMe

OAc.G-OH + H-G-NHMe

Ac-G-G-NHMe

H2O

COO

OHO COO

OOC

O

COO

OH

Chorismate mutase

Chorismate Prephenate

HN

N

O

O

R

COOH

HN

N

O

O

R

+ CO2

Orotidine 5'-phosphatedecarboxylase

3 Enzymes apparently can increase the rate of a chemical reaction by a factor of about 106 up to about 1017, compared to the same non-catalyzed process under the same conditions in aqueous solution !! For further examples see J. Am. Chem. Soc. 2000, 122, 11507; and Accts Chem. Res. 2001, 34, 938; also Annu. Rev. Biochem. 2011, 80, 645. The half-lives of various reactions that proceed spontaneously in water at 25oC in the absence of a catalyst have been estimated and compared (PNAS 2003, 100, 5607):

It is important to note, however, that many enzymes make use of either metal ions or other sall organic coenzymes, to catalyze reactions. By including the metal or coenzyme, reaction paths (mechanisms) become possible that would not be reasonable in their absence, e.g.

Rate constant at 25oC = 2 x 10-17 s-1. With PLP (but no enzyme) the reaction rate is greatly enhanced.

R COO

H3N H

R H

H3N H+ CO2

N

CHOOH

Me

O3PO2

4 Further interesting examples:

The following coenzymes play important roles in enzymic catalysis. Note that this is not a complete list of all known coenzymes:

NH3

COOACC Lyase

H2CCH2 CN CO2++

H3N N

COO ON COO

H

SH

O

H

O2,

α-Ketoglutarate,

Fe2+

Isopenicillin-

N synthaseN

S Me

Me

NH

O

H3N

OCOOH

COO

+ Succinate + H2O

NH

HN

NH

HN

NH

NHMe

OH

OH

OH

H2NOC

HO

O

O

O

O

OHO OH

Cl Cl

O

NH

HO OH

OOC

OOC N HO N

H

O

OHHO OH

HN

ON

O OH

H

O

CONH2

HN

ONH

O

ONH2Me

HOOHCl

Cl

3 x P450

O

O

O

OH

Me

Me

Me

Me

Me

OH

OH

MeMe

eryA

Propionyl-CoA +

6 Methylmalonyl-CoA

6-Deoxyerythronolide B

Me

O

SCoA Me

O

SCoA

COO

Vancomycin-Aglycon

Me

O

SCoA

Me

CO-SCoAIsobutyryl-CoA Mutase

Coenzym-B12

(Polyketid-Synthase)

+ 6 NADH

Me

MeMe

Me

OPP

Me

Me

Me

Me

MeTaxadiene-Synthase

H H

Me

5

Protohaem-IX(Electron-Transport.Oxygen Transport,Oxygen activation- P450s)

Biotin(Carboxylations)

Pyridoxal Phosphat(Aminoacid metabolism

Coenzyme- B12(Rearrangements)

X =

Lipoamide(Redox-Reactions)

Thiamin Pyrophosphate(Carbonyl-Umpolung)

Nicotinamide Adenine Dinucleotide (NAD+)(Redox-Reactions) Flavin Adenine Dinucleotide (FAD)

(Redox-Reactions)

S-Adenosylmethionine (SAM)(Methylation, amongst others)

Adenosine Triphosphate (ATP)(Phosphate transfer

Coenzyme-A (CoA-SH)(Acyl transfer reactions)

⊕

⊕N

CONH2

O

HO OH

O P O P O O

HO OH

N

N

N

N

NH2

O- O-

O O

N

N

N

N O

OH

Me

Me

CH2

CH-OHCH-OHCH-OHCH2

O PO

PO

O-O-

NH2

N

NN

N

OHHO

OOO

HOOC S

NH2

Me

O

HO OH

N

N

N

N

NH2

O P O P O O

HO OH

N

N

N

N

NH2

O- O-

O O

PO

-OO-

HSN NH

O

H

O

OH

Me Me

O PO

PO

OO

NH2

N

NN

N

OHHO

OOO

N+ S

N

N

NH2

Me Me OPO32-

H

SS N

O

Lys..ENZ

H

MeH2NOC

CONH2

X

Co

N Me

CONH2

CONH2

CONH2

CONH2

Me

N

NNMe

Me

Me

Me

Me

H

O NH

O

Me

PO

HOO

HO N

Co O

NMe

MeO

HO

HO OH

N

N

N

N

NH2

N

2-O3PO

CHO

OH

Me

HN NH

S

O

CONH Lys...ENZ

N N

N N

Fe

MeY

Z

Me

Me

COOHCOOH

Me

6 1.2. Mechanisms of Chemical Catalysis How does an enzyme (or any catalyst) accelerate a chemical reaction? Our explanation is usually based in transition state theory, i.e. that the rate of a chemical reaction is determined by the energy difference between the ground state and the transition state of the reaction (DG#). The transition state is treated as a thermodynamic entity in pre-equilibrium with the reactants, and whose concentration is proportional to the rate. The minimum energy pathway is called the reaction coordinate, and when plotted in 2D vs. energy gives the reaction coordinate diagram. The transition state is the highest point on the reaction pathway interconverting reactant and product:

Shown is the microscopic rate constant for a single step conversion (the Eyring equation), where krxn the reaction rate, T the temperature, κ is a transmission coefficient, k and h are Boltzmann’s and Planck’s constants, respectively, and ∆G‡ is the free energy of activation. The Arrhenius equation was derived empirically before the development of transition state theory, from the observation that reaction rates (the reaction may not be one-step!) increase exponentially as the temperature increases. So the two equations are not necessarily directly comparable.

𝑘 = 𝐴. 𝑒'()/+, When a catalyst is present, the reaction will follow a different mechanism with a lower activation free energy, compared to the process without the catalyst. It is also important to recognize that for any given reaction, the value of DGo is identical for both the catalyzed and non-catalyzed reactions; only DG# is smaller in the catalyzed proecess. Hence, both the forward and reverse reactions are accelerated by the catalyst, but the equilibrium constant remains unchanged. It is often very helpful in the analysis of the structures of transition states to remember the Hammond postulate. In the case of a strongly endothermic reaction step, the structure of the activated complex (transition state) resembles more closely that of the product, whereas in an exothermic reaction the structure of the transition state will resemble more closely that of the starting material. : From a chemical viewpoint this is reasonable:

E

Reaction

7 In the following discussion, we will start by looking at organic reactions that proceed without enzymic catalysis. Nevertheless, organic reactions can be catalyzed; but what sorts of catalyst are seen, and what are the mechanisms of the catalyzed reactions? 1.2.1. Specific-acid and specific-base catalysis There are two types of acid catalysis, and two types of base catalysis. Here we will deal with specific acid and specific base catalysis, or Bronsted acid-base catalysis, where in water solution a proton or hydroxide are involved in the reaction mechanism: Specific-acid catalysis involves catalysis by H+

(H3O+), and so the rate depends upon pH: In general terms: X + H3O+ XH+ + H2O products Rate = k [XH+] = k.K[H3O+].[X] In terms of reaction mechanism, a rapid equilibrium protonation of the molecule occurs in the first step, followed by a slow step, which is accelarated in comparison with the uncatalyzed reaction because of the greater reactivity of the protonated molecule. One well known example is ester hydrolysis ; Another example is the specific acid catalyzed hydration of acetone: The concentration of [XH+] depends upon the pH, and the concentration of X. This mechanism is therefore only important when pH ≤ pKa of XH+. Remember: HA + H2O H3O+ + A-

Keq = [H3O+] [A-] Ka = [H3O+] [A-] pKa = - log Ka [AH] [H2O] [AH] The pKa of HA corresponds to the pH where it is half dissociated in aqueous solution. In summary for specific-acid catalysis:

• H3O+ is the catalyst (is the only effective catalyst) • the protonation step is not rate limiting (fast pre-equilibrium) • a rate determining reaction of the protonated species • is effective only at or below the pKa of the protonated species. • the rate depends on pH • Usually only simple uni- and bi-molecular reactions. • Typically an inverse isotope effect is seen : k(H2O) < k (D2O)*

* D3O+ in D2O is a stronger acid than H3O+ in H2O. Water (H2O) is a better solvating agent for H3O+ than D2O for D3O+ because it forms stronger H-bonds.

8 In specific-base catalysis, the rate also depends on pH, i.e. on [OH-]. This typically involves a reaction with a first order dependence on [OH-]. It may involve the direct attack by OH- in the rate-determing step (RDS), or the rapid removal of a proton from the substrate by hydroxide ion (OH-) in a fast pre-equilibrium, followed by a rate-limiting reaction of the anion. For example:

Another example is the aldol reaction:

In D2O, D is rapidly incorporated into the aldehyde (occurs faster than the aldol reaction). In summary, for specific-base catalysis:

• HO- is the catalyst • the rate depends on pH • often means a rate-determining reaction of a deprotonated species • usually only simple uni- or bi-molecular steps

Specific-acid and specific-base catalysis are mechanisms that are not used by enzymes, since the pH of an enzymatic reaction inevitably must remain close to 7. 1.2.2. General-acid catalysis and general-base catalysis This is the second important type of catalysis by acids and bases, where the proton transfer is involved the rate-determining step (RDS), not in a prior equilibrium. Here, at constant pH, the rate depends on the concentration of a general-acid or a general-base. In such cases, the proton transfer is not complete before the rate-determining step, but occurs during the rate-determining step (RDS). e.g. in catalysis by a general-acid HA, or a general-base B :

An example of general-base (GB) catalysis: acetate accelarates the formation of esters from alcohols and acetic anhydride:

R SH + OHO

R

RDSO

RRS

OH

RRSRS +R SH

Me

O

H Me H

HO OO

H

Me

O

H

kobs

HO- / H2O

fast RDS+ H2O

KeqHO

HO+

H2O

krate

[HA]

krate

[B]

O

O

R

OH

O

O O O

OH RO

O

O

ORDS

9 Why does acetate accelarate this reaction ? Acetate (pKaH 4.7) is far too weak a base to deprotonate the alcohol (pKa ≈ 16). If it acted as a nucleophile there would be no net reaction. But acetate can remove a proton from the alcohol as the reaction occurs (in the RDS):

Acetate would be more than basic enough to deprotonate the intermediate, and at the transition state, the pKa's become matched, and the proton transfer occurs very rapidly:

A major disadavantage of this mechanism is that the RDS is termolecular, so the entropy of activation tends to be large and negative. This can be overcome if one component is in large excess (e.g. the alcohol is used as solvent). Another example:

Summary of general-base (GB) catalysis:

• At constant pH the rate is dependent upon the concentration of the general base. • proton transfer occurs at the transition state of the RDS. • is effective at neutral pH, even if below the pKa of the substrate (see above). • Ofter a termolecular RDS (large -DS#). • a normal solvent isotope effect (k(H)/k(D) ≈ 2-4) is seen, since an X-H (X = O, N) bond is broken

in the transition state of the RDS. In the case of general-acid (GA) catalysis, a reaction is catalysed by a general acid (HA a non-dissociated acid, but not H3O+). The general acid is usually not strong enough to protonate the substrate in a rapid pre-equilibrium before the RDS. Rather, the proton is transferred in the RDS (i.e. at the transition state). The rate of the reaction increases with increasing concentration of HA, at constant pH:

Me

O

O

O

Me

RO

H Me

O

O

O

Me

RO

H Me

OH

O

O

Me

RO

pKa (alcohol) >> pKaH of acetate pKa (intermediate) << pKaH of acetate

fast

O

O

R

OH

O

O O O

O O

O

O

ORDS

H

Rδ

δ

Transition state

HOCl O

ClRDS

HOCl

ArOO

H Cl

RDSO

HO

krate

[HA]

10 Normally, ester formation and ester hydrolysis require specific acid/base catalysis. The weaker general acid/base catalysis is not effective enough. However, in some special cases GA/GB catalysis is seen, e.g.

Normally, the formation and hydrolysis of acetals requires specific-acid catalysis, since alcohols are poor leaving groups; GA/GB catalysis is not very effective.

In some cases GA catalysis can be observed, for example, when there is additional stabilization of the carbocation intermediate:

In summary, in general-acid (GA) catalysis:

• any undissociated acid can function as a GA catalyst • the acid is too weak to protonate the substrate in a rapid pre-equilibrium • the proton is transferred in the rate-determining step (in the transition state) • this mechanism of catalysis can be effective even at neutral pH • the rate increases with increasing [HA], at constant pH • normal solvent kinetic isotope effects are seen • often termolecular with large -DS#

Another very interesting example where GA catalysis occurs during acetal hydrolysis (Tet.Lett. 1963, 4, 911-3):

O HO

O O OHOH

H

O O

O OR O ORH

O

OEt

OEt

AcOHOEt

H2Ofast

O

R

OMe

OMe

OMe

AcOHR

OMe

OMe R

OMe

OMe

H2Ofast

OOH

O

OO

H

OHHOHO

OOH

OOH

HOHO

O

OH

11 Concerted general-acid and general-base catalysis Concerted general-acid and general-base catalysis occurs infrequently except in enzymic reactions where it often occurs. Now an acid HA and a base B are both involved in the mechanism. As a result, such cases typically give bell-shaped rate vs. pH plots.

Non-enzymic model systems have been sought, but are difficult to find. One example (perhaps):

When acids and bases are not present, the solvent can have a large influence on the rate of mutarotation: Solvent. Rel. rate Benzene very slow Benzene + cat. H2O faster Pyridine very slow Cresol very slow Pyridin + Cresol fast and rate = k[glucose][cresol][pyridine] a-Pyridon best catalyst (although a weaker base than pyridine and a weaker acid than phenol)

A similar effect was not observed in aqueous solution.

log(kobs)

pHpKa of HA

General-acid catalysis

log(kobs)

pHpKa of BH+

General-base catalysis

log(kobs)

pH

combine

OHOCH2

HOHO

OH

OHOH

HOCH2

HOHO

OH

O

H

OHOCH2

HOHO

OHOH

ß-D-Glucopyranose[α]D = + 112o

α-D-Glucopyranose[α]D = + 18.7o

OHOCH2

HOHO

OH

O

H

N OH

12 Organocatalysis in asymmetric synthesis Organocatalysis has emerged recently as a new and powerful tool in chemical synthesis, in particular, the asymmetric version in enantioselective synthesis. Here, small organic molecules are designed to catalyze organic recations. The catalysts are often chiral, so enantioselective catalysis can be observed. One class of organocatalysts uses chiral phosphoric acids: Chiral phosphoric acids (Accts. Chem. Res. 2016, 49, 1029; ACIE, 2006, 45, 3909; ACIE 2008, 47, 4638) The aim is to design chiral acid catalysts. If these catalysts can accelarate a reaction by general-acid catalysis, then proton transfer in the RDS would occur in a chiral environment, and so in principle the reaction must be stereoselective.

Examples of nucleophilic addition to imines catalyzed by chiral phosphoric acids are shown below. Red indicates the functionality added to imine.

Examples of phosphoric acid catalyzed Strecker reactions:

R1

R2

O

OP

O

O H

R1, R2 =

iPr

iPr

iPr

F3C

F3C

CF3

SiPh3

chiral phosphoric acid catalysts

13 Other examples reported recently:

Friedel–Crafts reactions of indole and imines:

Possible transition state models have been investigated, which differ in the orientation of the bound electrophile and nucelophile within the cavity in the phosphoric acid catalyst. Which geometry is favored appears to depend upon the structures of the nucleophile and electrophile. For more details see: Accts. Chem. Res. 2016, 49, 1029. So far few detailed mechanistic studies on these catalysts have been reported. However, it seems clear that the chiral catalyst acts to protonate the electrophile within a chiral environment (rather like an enzyme !).

14 1.2.3. Covalent Catalysis - Nucleophilic Catalysis Here the catalyst does not simply act as an acid or base, but rather provides catalysis by acting as a nucleophile. The catalyst attacks the substrate and so opens a mechanistic path, which is not possible in the absence of the catalyst. As a result new intermediates will be generated on the reaction pathway, and in some cases these intermediates can be observed spectroscopically, or trapped, so providing evidence for the proposed mechanism. One example is the benzoin condensation:

Mechanism: CN- is a catalyst for this reaction because of its special properties; 1) it is a good nucleophile and attacks the carbonyl group rapidly, and 2) it can stabilize a negative charge in the a-position by resonance. Another well known example :

Here :

• The catalyst (pyridine) is a stronger nucleophile than the alcohol in the non-catalysed reaction (without pyridine).

• The pyridinium group in the intemediate is a much better leaving group, and • This intermediate is less stable than the product (the ester).

The powerful catalytic effects of 4-(dimethylamino)pyridine (DMAP) in acyl transfer reactions is well known, and DMAP is frequently used for this purpose (see Angew. Chem. Int. Ed. 2004, 43, 5436). Chiral amines in organocatalysis The use of chiral secondary amines to catalyze a wide variety of organic reactions enantioselectively has grown enormously in importance in recent years (for reviews see: Drug Discovery Today 2007, 12, 8-27; ACIE 2008, 47, 4638). One of the first reported examples is the enantioselective aldol reaction catalysed by the amino acid proline:

The mechanism has been well studied, and begins with the 2o-amine acting as a nucleophilic catalyst to generate an iminium-ion. Iminium-ions are much more reactive than neutral imines; they are formed and can be hydrolyzed more rapidly, and can be rapidly and reversible converted into enamines. An aldol-like

Ph CHO Ph C

O

CH

OH

PhCN-, H2O

Me O Me

O OR OH

Me OR

O+ AcOH

Pyridine+

Me Me

O

H

O

R

NH

COOHcatalyst(30mol%)

DMSOor DMF

R

O OH

54-97% yield60-96% eeR = iPr, Ph, 1-naphthyl

15 reaction can then occur, with the carboxylic acid acting as a general acid catalyst:

Or in the general case:

Some related proline-catalyzed reactions reviewed in Drug Discovery Today 2007, 12, 8-27:

Me Me

O N COOH

H N

Me

O

OH

O

R H

N

Me

O

OH

Me

16 Different organocatalysts have also been developed that follow a different stereochemical course:

The simple steric models shown below rationalize a wide variety of reactions catalyzed by proline (I) and diarylprolinol ethers (II):

1.2.4. Covalent catalysis - Electrophilic catalysis Here the catalyst functions as an electrophile by withdrawing electron density from the reaction center. In general, electrophilic catalysts may include Lewis and Bronsted acids, metal ions, as well as other organic "cofactors" (e.g. coenzymes like pyridoxal phosphate). Two important effects of metal ions are : An interesting model reaction was reported several years ago, which demonstrates the catalytic effects of metal ions on amide hydrolysis (J. Am. Chem. Soc., 1993, 115, 1157):

17

The hydrolysis of phosphate monoesters is especially slow in water. However, phosphatases can catalyze this reaction, and are some of the most efficient enzyme catalysts known. How do they do this ? This is still a topic of intense research. But some phosphatases are known to possess an active site containing two metal ions. The structure of the ligated metal ions in the purple phosphatase I (PAP) is known from X-ray crystallography. Model complexes have also been synthesized with similar structures : Active site in PAP

The catalyst shown left can catalyze the hydrolysis of p-nitrophenylphosphate by a factor of 1012. The alkaline phosphatase superfamily catalyze phosphate monoester hydrolysis using a common mechanism with two Zn2+ ions bound at the active site (JACS 2016, 138, 14273):

"Organic" electrophilic catalysis Organic groups that can act as a p-acceptor, may also be classified as electrophilic catalysts. For example the decarboxylation of ß-ketoacids is catalyzed by amines :

We have already seen that amines can catalyze aldol reactions (and retro-aldol reactions). Catalytic mechanisms such as these play a very important role in many enzymic reactions.

Co3+

NH

NHHN OH2

HN OH2Co3+

NH

NHHN OH-

HN OH2Co3+

NH

NHHN OH-

HN OH

N

O

HNOH

O

OH

60oC

Vgl. HO- kat. Reaktion : k = 9 x 10-12 s-1 at pH 5.9 and 60 oC

+k = 3.6 x 10-4 s-1

NHCo

OH

OH

O

NHNH

CoNH

O

NHNH

POOO2N

3 3

His325Fe

OH

O

O

Asp135Tyr167

ZnHis323

O

His285Asn201

POO

O Asp164

3 2

Me O-

Me

O

Me Me

O

O-

N

Me Me

OR H

kkat

kunkat

⊕R-NH2

18 1.2.5. Preorganisation When 2 molecules (A + B) react with one another, they first have to collide and associate, and for this an entropy barrier must be overcome before they can cross the transition state (TS). Enzymic reactions are fundamentally different, in that the first step is always the binding of the substrate(s) (S) at the active site of the enzyme (E), to form a so-called Michaelis complex (ES). Only after the substrates have bound (and perhaps after complex conformational changes have occurred in the enzyme) does a reaction take place. The entropic barrier, therefore, is now encountered in the formation of the ES complex, and not in the process leading to the TS. The reaction at the active site, therefore, is rather analogous to an intramolecular process, in which all the participating functional groups are not only collected together, but also preorganized for a reaction to occur, i.e. bound in the correct conformations, and in the correct relative positions and orientations, both with respect to groups in the substrates but also catalytic groups in the enzyme. It is even conceivable, that the atomic motions that have to occur along the reaction pathway can be coordinated with the natural breathing (i.e. dynamic motions) of the protein. But in terms of catalytic efficiency, how much are these effects worth; how much do they contribute? This is a fundamentally important question that has still not been completely answered. However, some important insights have come from studies of simpler model systems, in particular, from comparisons of inter- vs. intra-molecular reactions.

For example, intramolecular reactions can be considered as models for enzymic reactions, in which the reacting groups are held together, and perhaps also correctly oriented for the reaction to occur. Such “proximity-effects” can have a dramatic influence upon reaction rates :

+

+

E

k

Ke kcat

k

random collision

preorganizationdesolvation

preorganisationin a model system

bimolecular reaction

enzymatic reaction

intramoleculare reaktion

A B

AAB B

A B

A B A B

A B A BA B

K

K

19

Before a nucleophilic attack can occur, the attacking nucleophile must approach from the correct position and trajectory; this is strongly influenced by the preferred conformation of the molecule: Near attack conformations (NACs) are defined as conformations that achieve the optimal arrangement of juxtaposed reactants to proceed to the transitions state. The rate constants for several intramolecular reactions directly correlate to the mole-fractions of the reactants present as NACs. The efficiency of intramolecular catalysis can be estimated with the so-called "effective molarity" - The ratio of the first-order rate constant of an intramolecular reaction involving two functional groups within the same molecular entity, to the second-order rate constant of an analogous intermolecular elementary reaction. This ratio has the dimension of concentration. In other words, the EM-value corresponds to the concentration of the catalytic group required in the intermolecular process in order that it proceeds as fast as the equivalent intramolecular process (Adv. Phys. Org. Chem. 1980, 17, 183).

In early studies it proved difficult to find model reactions with EM values larger than 10-100 M, wherein a proton transfer (and not a nucleophilic attack) was involved in the RDS. Does this mean that acid catalysis is not so important ?

CH3COO- + CH3COOC6H5Br (p)

COOC6H4Br(p)

COO-

COOC6H4Br(p)

COO-Et

Et

COOC6H4Br(p)

COO-

COOC6H4Br(p)

COO-

COOC6H4Br(p)

COO-Ph

PhCOOC6H4Br(p)

COO-iPr

iPrO

COOC6H4Br(p)

COO-

krel

1.0

1x103

3.6x103

1.8x105

2.3x105

2.7x105

1.3x106

8x107

COO-

O

O NO2

Me COO- Me

O

O NO2

H2O

COOHOH

COOHOH

COOHOH

COOHOH

COOHOH

EM (M) for lactone formation

4 x 104 7 x 106 2.3 x 107 3.7 x 1011 2 x 1013

20 In later studies, however, much larger EM values were observed in systems where efficient intramolecular proton transfer occurs, and a feature of such cases is a strong H-bond (preorganization) in the reactant and/or product:

(Kirby & Williams J. Chem. Soc. Chem. Comm. 1991, 1643).

(J.Chem.Soc. Chem.Comm., 1994, 707; Accts. Chem. Res. 1997, 30, 290-6) The formation of a strong hydrogen bond in the product appears to characterize the high EM values obtained in these systems. They also indicate that very efficient proton transfer occurs in the TS. Finally, such efficiency in the proton transfer step seems to require a precise positioning (preorganization) of the reactive groups (see Angew. Chem. Int. Ed. 1996, 35, 707). This very precise positioning of reactive groups is something that can be anticipated in an enzymic reaction, where the spatial interactions between functional groups can be precisely controlled by substrate binding at the active site. 1.2.6. Electrostatic effects Water possesses a rather high dielectric constant e = 79. The electrostatic interaction between two point charges e1 and e2 depends on their distance appart, and on e : E = e1 . e2 e.r Water as solvent reduces dramatically the energy of electrostatic interactions : Electron ° ° Proton r = 3.3Å in Vacuo ( e =1) E = -100 Kcal/mol in water ( e =79) E = -1.3 Kcal/mol The interior of a protein is a very heterogeneous environment, where the dielectric constant is certainly

O

OR'R

O

OH

OR'

O

OH

O

OR'R

O

OH

δ

δ

OR'

O

O

OMeO NMe2H

O NMe2H

pKa (OH) = 14.9

MeO NMe2H

MeO NMe2H

NMe2

OMe

21 smaller than in bulk water. So electrostatic interactions will be more important at the active site of an enzyme than in water. Unfortunately, it is rather difficult to exactly measure (or calculate) the precise dielectric in the interior of a protein. But the importance of such effects must not be overlooked, e.g. what is the energetic value two H-bonds ? -> depends on the environment !

Not general acid catalysis !! Enzymes can also use the protein architecture to impose specific electrostatic fields onto their bound substrates, but the magnitude and catalytic effect of these electric fields have proven difficult to quantify. One interesting case is the active site of the enzyme ketosteroid isomerase, which exterts an extremely large electric field onto a carbonyl C=O bond, thereby increasing catalysis (Science 2014, 346, 1510). 1.3 Enzyme kinetics Although enzymic reactions take place in aqueous solution, the substrate is not transformed into product in free solution, but rather whilst bound at the active site of the enzyme. So the substrate must bind to the active site of the enzyme before a reaction can be catalyzed. Note that the process of substrate binding quite often causes large-scale conformational changes in the enzyme itself, such as the closure of loops over the active site, which can for example exclude water from the vicinity of the active site during catalysis. To a first approximation, the structure of the active site should be complementary (i.e. bind tightest) to the transition state of the reaction:

In the simplest case: We are often first interested in so-called steady state kinetics, which represents the changes in product concentration (the initial rate) during the first few seconds-to-minutes of the reaction (i.e. <10% of the substrate has reacted). An important assumption is that [S] >> [Eo], so that the concentrations of all enzymic species can be neglected. If this is not the case, the kinetic treatment is more complicated. However, under these conditions ([S] >> [Eo]), at any given enzyme concentration, the rate (v) depends on [S] and the parameter Km (Michaelis constant), as shown below:

O

OSerNH

R

H H

NN

O

O

Enzyme Substrate

+

ES-complextransition state

22 This type of kinetic behavior is described by the Michaelis-Menten equation: v = kcat [Eo] [S] (Michaelis-Menten equation) Km + [S] Thus, as [S] becomes larger and finally [S] >> Km, then: v = kcat [Eo] = Vmax (often just written as V) Also: v = Vmax [S] when [S] = Km then v = 1/2.Vmax Km + [S] Note also when [S] << Km the Michaelis-Menten equation can be simplified : v = kcat [Eo] [S] ≈ kcat [Eo] [S] Km + [S] Km What is the significance of kcat, Km and kcat/Km ? 1. kcat is a first order rate constant (s-1). This is called the catalytic constant or turnover number. It corresponds to the maximum number of substrate molecules that can be converted into product per second per enzyme molecule. Typical values are in the range from 1-105 s-1. In order to measure this constant, the enzyme concentration must be known. kcat is the sum and weighted property of the ES and all other E-intermediate complexes up to the EP complex, that are formed along the reaction path. Only in the simplest case (see below) can kcat be identified with one specific step in the reaction path. In the most general case, it describes how quickly all enzyme-bound species are converted into free product (P). If one of the steps is much slower than all the others, then this will dominate the kcat value. Of course, kcat cannot be larger than any one of the first order steps along the reaction pathway (k1, k2 etc.). 2. The Km is the substrate concentration needed to achieve half-maximal velocity, i.e. 1/2 Vmax. Typical values are in the range 10-1-10-7 M. The Km value describes how tightly all enzyme-bound species interact with the enzyme, or in other words, it can be viewed as a global dissociation constant for all enzyme bound species. A high value means weak binding, a low value means tight binding. Again, if there is one intermediate that binds much more tightly to the enzyme than all others, this will have a higher weighting in determining Km. 3. As shown above, kcat/Km is itself an important constant that is equivalent to the second order rate constant for the reaction between E and S (s-1 M-1). This constant is a property of the free E and the free S. It is frequently used as a measure of the efficiency or specificity (when comparing two or more substrates) of an enzyme. Typical values are 106-109 M-1 s-1. At low [S], then [E] ≈ [Eo] so : v = kcat [E] [S] Km Thus, when [S] << Km the ratio kcat/Km is an "apparent" second order rate constant (M-1 s-1) for the enzymic reaction. But note that the value cannot be higher than the diffusion-controlled rate of a reaction in aqueous solution (the upper limit is ca. 109 M-1 s-1).

23 We can represent these constants on one typical reaction energy diagramm:

How was the Michaelis-Menten equation derived (see Biochemistry, 2011, 50, 8264)? Important contributions to the theory of enzyme kinetics were made by A. J. Brown und V. Henri (Univ. Zürich). However, it is usually the later contributions of Michaelis and Menten that are taken as a starting point in discussions of enzyme kinetics (for an historical account: FEBS Lett 2013, 587, 2725 & 2753). Michaelis and Menten assumed that the first step was a rapid equilibrium, and the substrate concentration is in excess over the enzyme concentration (i.e. so that total substrate conc. [So] ≈ [S]), where then: KS k (=kcat) E + S ES E + P concn (Eo - x) S x where Eo = total enzyme conc. [E][S] [Eo - x] [S] x = [Eo] [S] so the rate v = k . x = k. [Eo] [S] Michaelis and Menten assumed that many enzymic reactions could be described by this equation, and that the formation of ES would be rapid compared to the subsequent "chemical" step. Later Briggs and Haldane suggested that it was not necessary for the binding of substrate to be a rapid pre-equilibrium step. Rather, they assumed that a steady state was quickly established, in which the concentrations of all forms of E did not change during the initial phase (when [S]>>[E]), and their concentration was determined by how fast they were formed and how fast they break down. The steady state assumption made by Briggs and Haldane is a more general mechanism in which the binding of S is not assumed to be reversible, nor necessarily at equilibrium during the reaction: k1 k2 E + S ES E + P

E + S

ES EP E + PKs

kcatKm kcat

E

Reaktion

[ES]

KS = [x]

=

KS + [S]

KS + [S]

24 k-1 concn (Eo - x) S x Briggs and Haldane postulated that very rapidly (within milliseconds of mixing everything) a steady state would be established (again, when [S] >> [E] should be true), where: dx k1 [Eo - x] [S] - k-1[x] - k2 [x] dt Rearranging: x = k1 [Eo] [S] k-1 + k2 + k1 [S] so v = x k2 = k1. k2 [Eo] [S] k-1 + k2 + k1 [S] or, dividing all terms by k1 : v = kcat [Eo] [S] where kcat = k2 and Km = k-1 + k2 Km + [S] k1 This again is the Michalis-Menten equation. How can Vmax, kcat und Km be determined? The Km und Vmax (or just V) are determined by measuring the initial rate (v) as a function of the concentration of one substrate ([S]), while keeping all other concentrations constant, and then repeating this at several additional fixed levels of a second substrate (if required). The data are then displayed graphically using linear transforms of the Michaelis-Menten equation:

Nowadays, a lap-top computer and a commercial software package can be used to directly fit by non-linear regression the experimental data to the Michaelis-Menten equation to determine the constants Km and kcat. Note, however, that in order to determine kcat the value of [Eo] must be known, and this assumes that the enzyme is available in pure form and that its concentration can be measured. This is not necessary when the aim is to measure Km and Vmax - this can be done without knowing the value of [Eo], and so can also be done using crude cell extracts containing the enzyme activity of interest.

V . [S][S]

V V[S]

V V

V

v =kcat [Eo] [S]Km + [S]

=V [S]

Km + [S]

1v

= 1V + Km oder

v= Km +

1v

1[S]

- 1 Km Grad = Km

[S]v

[S]

- Km Grad = 1

Km

= 0 =

25 1.4 The role of general acid-base catalysis and nucleophilic catalysis in enzymic reactions: glycosidases and glycosyl-transferases Glycosidases catalyze the hydrolysis of oligosaccharides. But how stable are glycoside derivatives under physiological conditions? The rate of hydrolysis of ß-methyl-D-glucopyranoside has been investigated, and compared to that of some other molecules related to important biopolymers (J. Am. Chem. Soc. 1998, 120, 6814): Arrhenius plot for hydrolysis at pH 7

When compared to the rate of phosphodiester hydrolysis (oligonucleotides) and amide bond hydrolysis (peptides and proteins), the rate of glycoside hydrolysis under comparable conditions is slower: oligosaccaharides are the most stable of the three biopolymers. Glycosidases are very efficient catalysts. They can accelarate the rate of oligosaccharide hydrolysis by factors up to 1017 compared to the rate of spontaneous hydrolysis under comparable (physiological) conditions. The glycosidases are a very large class of enzymes, with over 2000 from different origins now known. These enzymes can be divided mainly into two large classes, that have different mechanisms of action, and different stereochemical courses (see: Curr. Opin. Chem. Biol., 2000, 4, 573; Accts. Chem. Res. 2000, 33, 11): One family is called the “inverting glycosidases” - they catalyze hydrolysis with inversion of configuration at the anomeric center:

Here the distance between the precisely positioned general acid and general base in the enzyme's active site is optimal (~10.5Å) to allow a more-or-less concerted hydrolysis mechanism. In such cases no

OO R

HO O

O H

H

O O

O OR

HO O

O H

HO O δ

δ

δδO

OR

H

26 covalently bound E-S intermediates have been detected. This is not the case with the second major family, the retaining glycosidases, which proceed with retention of configuration at the anomeric center: The “retaining glycosidases” - Here the two sides of the active site are only ~5.5Å apart:

An important difference is that a covalent enzyme-intermediate complex is now formed. Support for this mechanism can be obtained by demonstrating that this intermediate is indeed formed. But how ? Through subtle modification of the substrate this was possible by X-ray crystallography (Vgl. Biochemistry 1998, 37, 11707): X-ray crystal structures

replacing 2-OH by -F leads to:

• loss of H-bonding interactions in the active site, which are important in TS stabilization, and • inductively destabilizes an oxo-carbenium ion-like TS, so breakdown of the intermediate is slow.

HO O

O O

O OR

HO O

O O δ

δ

δδ O

O O

O O

OH

H

O OH

HO O

O O δ

δ

δδOO

H

HO O

O O

OO R

H

27

The first crystal structure of an enzyme was that of lysozyme, first published in 1965. However, only in the past few years was the mechanism of the lysozyme reaction proven. This enzyme catalyzes the hydrolysis of peptidoglycan, an important component of the cell walls of bacteria. Lysozyme therefore kills bacteria and is produced to help defend organisms from bacterial attack. The structure and biosynthesis of peptidoglycan are shown below:

Molecules that block (inhibit) the biosynthesis of peptidoglycan act as antibiotics. Well known examples include penicillin und vancomycin:

The crystal structures of various complexes formed between lysozyme and oligosaccharide fragments showed how the substrate binds at the active site of the enzyme. These structures also revealed two key groups in the enzyme that are essential for catalysis Glu35 und Asp52, which lie on opposite sids of the substrate binding cleft. One (Glu35) exists in a neutral (undissociated) form, and the other (Asp52) as a carboxylate anion :

P

P-

-P

P-

-

H

crosslinkedpeptidoglycancell wall

vancomycin inhibitstransglycosylationL-Lys-D-Ala-D-Ala

OO

NHAcO

Me

OHN

O

OH

L-Ala

γ-D-Glu

L-Lys

D-Ala

D-Ala

O

NHAcHOO

OH

γ-D-Glu

L-Lys

D-Ala

D-Ala

NH2

OOCN H

O

HN O

OHHO OH

HN

ON

O

H

O

CONH2

HN

ONH

O

ONH2Me

HOOH

Cl

ClOO

OH

OHOH

OH2N

Me

HOMe

Vancomycin

N

S

O

Me

Me

HN

O

R

COOH

H

Penicillins

28

The reaction proceeds with retention of configuration at the anomeric center. For a long time, there was great uncertainty about the mechanism, because a covalent ES complex (as intermediate) had never been detected. Either a carbocation is formed which reacts further to product, or a covalent intermediate is generated:

Only recently were experiments described, which proved that a covalent ES complex can be formed. In order to achieve this the chemical properties of the substrate were altered to allow the intermediate to accumulate; Glu35 was mutated to Gln35 (deactivates the enzyme) and the substrate used was a glycosyl fluoride (more reactive):

The activated glycoside substrate can still react, but breakdown of the covalent intermediate is slowed by a factor of 105 (half life now 64 h) due to destabilization of the oxo-carbenium ion-like TS by 2-F.

O OO

OH

OO NHAc

NHAcHOO

OH

O OH

OH

OO NHAc

HOO

NHAcHOO

OH+

NAM NAG

OOR

H

OH

ROHO

NHAc

O H

OH

ROHO

NHAc

⊕

O H

O

OH

ROHO

NHAcO

Asp52

OOC Glu535

OO

H

Glu35

O O

Asp52

OO

Glu35

OH

H

OOC Asp52

O F

H

OH

ROHO

X

O H

O

OH

ROHO

XO

Asp52

ONH2

Glu35

O O

Asp52

X = NHAc = F

29 In this way, the concentration of the stationary concentration of the intermediate could be raised enough to allow detection by MS and X-ray crystallography (Nature, 2001, 412, 835). Inhibitors of glycosidases are also important in medicinal chemistry. For example, neuraminidase (NA) is a membrane bound enzyme on the influenza virus, which is important for the life cycle of the virus. NA catalyzes the removal of terminal sialic acid linked to glycoproteins and glycolipids on the cell surface. The sialic acid is the ligand on the cell surface that binds to the other major protein on the influenza virus, the hemagglutinin (HA). The reaction catalyzed is:

influenza virus NA appears to be crucial in the release of progeny virion from infected cells, and in the movement of virus through the mucus of the respiratory tract, thus reducing the propensity of the virus to aggregate. Two inhibitors of NA are Tamiflu and Relenza, which each bind with high affinity to the active site of the enzyme because they mimic the geometry of the transition state in the hydrolysis reaction: The inhibition of enzymes active in carbohydrate metabolism is also valuable in the treatment of diabetes. For example, various glycosylamines are inhibitors of a-glucosidases used to digest carbohydrates. They are all currently used to treat type-2 diabetes (e.g. acarbose, miglitol, voglibose). Also of great importance are glycosyl transferases, which add sugar residues to (for example) proteins (to make glycoproteins) or small molecule natural products, e.g. antibiotics (see later). Glycosylation is the most frequently observed type of post-translational modification of proteins. Glycosyl transferases are therefore of great interest in biotechnology.

O

COOH

O

OH

AcNH

HO

HO

OH

sialic acid(N-acetylneuraminic acid)

O

COOH

OH

OH

AcNH

HO

HO

OH

NA

COOEtO

NH2.H3PO4

AcNH

Tamiflu (Roche)(Oseltamivir phosphate)

O COOHH

HO

OH

HO

AcNHHN

NH

NH2

Relenza (GSK)Zanamivir

30 Activated sugar residues are required as glycosyl donors (not free sugars), and one typical example proceeding with inversion (a -> ß) is shown below:

O O

O

OP

O

O

O PO

O

O ON

HO OH

NH

O

OO RH

Mn2+

invertingα to ßglycosyltransferase